Efficient metal processing using high average power ultrafast laser

Size: px

Start display at page:

Download "Efficient metal processing using high average power ultrafast laser"

Nickolas Fisher
5 years ago
Views:

1 Efficient metal processing using high average power ultrafast laser Strasbourg, September 13 th, 2017 J. Lopez J. Lopez et al., Journal of Laser Micro and Nanofabrication, submitted (2017) G. Mincuzzi R. Kling K. Mishchik E. Audouard C. Hönninger E. Mottay Motivations Efficient metal processing using high average power ultrafast laser John Lopez September 13th

GF Machining Solutions Rollertech Group Motivations Increasing interest

embossing, printing) Texturing or engraving (jewelry, watch,

(tiny spot) Large area to process Short pulse laser : limited

pulse laser : promising technology enables to mitigate the thermal load

2 GF Machining Solutions Rollertech Group Motivations Increasing interest for high throughput surface processing : Manufacturing tools (molding, embossing, printing) Texturing or engraving (jewelry, watch, automotive) Hard to produce : High quality issues (no HAZ) & accuracy (tiny spot) Large area to process Short pulse laser : limited productivity due to uncontrolled melting (10 mm 3 /min) Ultrashort pulse laser : promising technology enables to mitigate the thermal load Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Scaling up capability? How to scale-up the ablation process? Increasing average power, fluence, rep. rate, overlap, scanning velocity? How to take advantage of new industrial tools?

3 Scaling up capability? How to scale-up the ablation process? Increasing average power, fluence, rep. rate, overlap, scanning velocity? How to take advantage of new industrial tools? Ultrafast laser : up to 100 W & 20 MHz Beam deflector devices : Galvo & Polygon The purpose of this study is to identify parameters of influence and process windows for surface ablation with a 100W / 10MHz ultrashort pulse laser combined with galvo or polygon scanners Efficient metal processing using high average power ultrafast laser John Lopez September 13th

4 1. Motivations Outline 2. USP ablation mechanism i. Kinetic of laser ablation ii. Heat accumulation at high fluence & high rep. rate 3. Experimental i. Setup & ultrafast laser ii. Protocol & sample analysis 4. Results & discussion i. Galvo scanner ii. Polygon scanner iii. How long to remove 1 mm 3? 5. Summary Efficient metal processing using high average power ultrafast laser John Lopez September 13th

measured on 40x40µm² Low fluence : high quality, low removal rate High fluence : poor quality, high removal rate C.

5 Kinetic of laser ablation ǀ Role of fluence Ablation depth vs fluence High fluence -> thermal load Thermal Regime 0.5 J/cm² 200 fs 200 J/cm² Optical Regime Ablation Threshold Parameters : Copper in vacuum, 150 fs, 780nm Mean depth is measured on 40x40µm² Low fluence : high quality, low removal rate High fluence : poor quality, high removal rate C. Momma et al., Appli. Surf. Sci. 109/110, (1997) Efficient metal processing using high average power ultrafast laser John Lopez September 13th

6 V/ t [mm3 min -1 W -1 ] Optimum in fluence Specific Removal Rate vs Fluence Max. There is an optimal fluence which depends on the pulse duration High fluences are less efficient / th Optimal fluence 2 Neuenschwander et al, Phys. Proc., 41, (2013) Neuenschwander et al, Phys. Proc. 56, (2014) Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Number of pulses to drill through [ x 1000 ] Heat accumulation induced by high repetition rate

Rate -> thermal load 100 50 40 µm 100 khz Stainless Steel 19ps 50µJ 300 khz 0 Rising rep.

ablation efficiency Heat accumulation Improves ablation efficiency but enhances HAZ Ancona et

, Vol 34, No 21, 3304-3306 (2009) Higher drilling efficiency for shorter pulses Heat

7 Number of pulses to drill through [ x 1000 ] Heat accumulation induced by high repetition rate Pulse duration: 19 ps 6 ps 800 fs Stainless steel 30µJ/pls High Rep. Rate -> thermal load µm 100 khz Stainless Steel 19ps 50µJ 300 khz 0 Rising rep. rate Repetition rate [ khz ] Particles shielding Reduces ablation efficiency Heat accumulation Improves ablation efficiency but enhances HAZ Ancona et al., Opt. Lett., Vol 34, No 21, (2009) Higher drilling efficiency for shorter pulses Heat accumulation overbalances particles shielding at khz High rep. rate enhances detrimental side-effects Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Influence of heat accumulation on surface morphology Laser irradiation introduces residual heat into

Saturation Temperature For a critical value between 600 and 800 C on Steel the surface morphology

4301 Stainless Steel Smooth & reflective surface Bumpy & oxidized surface No melting Bauer and al.

8 Influence of heat accumulation on surface morphology Laser irradiation introduces residual heat into the target. If subsequent pulses are too close in time or distance, the surface temperature raises. Saturation Temperature For a critical value between 600 and 800 C on Steel the surface morphology changes Simulation of Heat Accumulation 6ps 1030nm 800kHz 0.37J/cm² 4m/s, on Stainless Steel Smooth & reflective surface Bumpy & oxidized surface No melting Bauer and al., Optics Express, Vol. 23, No. 2 (2015) Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Remaining questions? Can we take advantage of 100W? How? P av = E pulse x RR Low fluence only Where is the limit? Limitations / process windows with galvo or polygon scanners?

9 Remaining questions? Can we take advantage of 100W? How? P av = E pulse x RR Low fluence only Where is the limit? Limitations / process windows with galvo or polygon scanners? Correlation between ablation efficiency, multipass capability and surface morphology? Efficient metal processing using high average power ultrafast laser John Lopez September 13th

10 Laser USP Laser & Experimental setup Pulse Picker Tangor 100 W 1030 nm <500fs 100 khz to 10 MHz M² 1.2 /2 Energy controller Cube Zoom x2 Beam expander Zoom x 3 Laser Laser Galvo head Scanlab Intelli Scan III-14 f(θ)100mm Spot 30 ± 2µm up to 5 m/s Steel foil AISI316L 500µm-thick Work piece on XYZ stages Nextscan LSE170 f(θ)190mm Spot 38 ± 2µm up to 100 m/s Beam expander Polygonal scanner Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Experimental protocol Laser Irradiation Laser

Scanning Electron Microscopy f(θ)100mm Spot

or multipass (30, 60, 120, 240) Velocity : 0.

Power : 10 to 100 W Lopez et al, SPIE, Vol.

Efficient metal processing using high average

11 Experimental protocol Laser Irradiation Laser 500 fs Confocal Microscopy for Topography Scanning Electron Microscopy f(θ)100mm Spot 30 µm Steel foil Process parameters : Single or multipass (30, 60, 120, 240) Velocity : 0.1 to 100 m/s Rep. rate : 0.5 to 10 MHz Ave. Power : 10 to 100 W Lopez et al, SPIE, Vol (2013) Lopez et al, ICALEO, M303 (2013) Efficient metal processing using high average power ultrafast laser John Lopez September 13th

12 Ablation efficiency Ablation efficiency is defined by the ratio : ρ = E th = C p (T)* T Sol->Liq + H melt + C p (T)* T Liq->Gas + H boil Energy required to bring 1 mm 3 from solid to gas with calculation based on thermo dynamical laws E th is given by Shomate equation, ρ = 1 ρ < 1 E th E exp 1 mm 3 Lopez et al., JLMN., vol. 10 (2015) Lopez et al., JLA, vol. 27 (2015) P av E exp = S * V Energy required to engrave a 1 mm 3 groove in our experiment Ideal process : all energy is used for ablation Real process : part of energy is lost (heating, shielding, scattering) Efficient metal processing using high average power ultrafast laser John Lopez September 13th

13 Etch rate [µm 3 pls -1 ] Specific Removal Rate [mm3ˑmin -1ˑW -1 ] Previous work / Influence of pulse duration Etch rate / Removal rate Specific Removal Rate Fluence [J cm -2 ] Lopez et al., JLA, vol. 27 (2015) Lopez et al., JLMN., vol. 10 (2015) Exp. data Spline fit Fluence [J cm -2 ] High fluence induces high throughput but low efficiency Specific Removal Rate decreases with increasing pulse duration Efficient metal processing using high average power ultrafast laser John Lopez September 13th

75 Groove collapses due to melting 90 W 36 W Depth 14 µm 15 mm 3 /min Eff. 0.

ablation efficiency above 2 J/cm² BUT introduces detrimental thermal effects 500fs 2MHz

14 Ablation efficiency [a.u.] Ablation efficiency versus fluence Galvo 1m/s Depth 40 µm 53 mm 3 /min Eff Groove collapses due to melting 90 W 36 W Depth 14 µm 15 mm 3 /min Eff Burr (10µm) Data 2017 Data W Fluence [J cm -2 ] Heat accumulation enhances ablation efficiency above 2 J/cm² BUT introduces detrimental thermal effects 500fs 2MHz 1m/s pitch 0.5µm overlap 98.3% Depth 1.2 µm 1.6 mm 3 /min Eff Smooth 40 µm Efficient metal processing using high average power ultrafast laser John Lopez September 13th

15 Ablation efficiency [a.u.] Multipass processing capability with Galvo 9 W 500fs 5m/s Average Power 9 W 40 µm Number of passes Efficient metal processing using high average power ultrafast laser John Lopez September 13th

16 Ablation efficiency [a.u.] Multipass processing capability with Galvo 9 W 500fs 5m/s Average Power 9 W 40 µm Average Power 45 W 45 W Number of passes Number of passes No multipass capability at 45 W & 90 W Efficient metal processing using high average power ultrafast laser John Lopez September 13th

17 Ablation efficiency [a.u.] 500fs 5m/s Multipass processing capability with Galvo Average Power 9 W 9 W 40 µm Average Power 45 W 45 W Number of passes Average Power 90 W 90 W Number of passes Number of passes No multipass capability at 45 W & 90 W Efficient metal processing using high average power ultrafast laser John Lopez September 13th

18 2D multipass processing / Cavity Cavity Dimensions 2.5 x 1 mm² Pulse-to-pulse pitch 1 µm Line-to-line pitch 1 µm Power [W] Fluence [J/cm²] Nb Pass Depth [µm] Laser Cavity ablated Ablation Efficiency Etch rate [mm3/min] fs 2MHz 2m/s pitch 1µm overlap 96.7% Single Line Dimensions 2.5 mm Pulse-to-pulse pitch 1 µm Cavity Single Line Etch rate [mm3/min] 2.34 Laser Groove Efficient metal processing using high average power ultrafast laser John Lopez September 13th Similar etch rate measured for single line and cavity - -

2D multipass processing / Cavity Cavity Dimensions 2.5 x 1 mm² Pulse-to-pulse pitch 1 µm Line-to-line pitch 1 µm Power [W] Fluence [J/cm²] 9 0.64 45 3.

6 J/cm² Dimensions 2.5 mm Pulse-to-pulse pitch 1 µm Cavity Ra 0.5 µm Single Line Etch rate [mm3/min] 2.34 3.2 J/cm² - - 4.

19 2D multipass processing / Cavity Cavity Dimensions 2.5 x 1 mm² Pulse-to-pulse pitch 1 µm Line-to-line pitch 1 µm Power [W] Fluence [J/cm²] Nb Pass Depth [µm] Laser Cavity ablated Ablation Efficiency Etch rate [mm3/min] Single Line 0.6 J/cm² Dimensions 2.5 mm Pulse-to-pulse pitch 1 µm Cavity Ra 0.5 µm Single Line Etch rate [mm3/min] J/cm² Similar etch rate measured for single line and 120 cavity µm - - Laser Groove 500fs 2MHz 2m/s pitch 1µm overlap 96.7% Ra 6 µm Efficient metal processing using high average power ultrafast laser John Lopez September 13th

20 Influence of scanning velocity at 10 MHz 5 m/s 4 m/s 3 m/s 2 m/s 40 µm Acceptable quality Eff m/s Burrs appear Eff m/s Overthickness 500fs 10MHz 50W 0.3J/cm² Melting flow appears Eff m/s Groove starts to collapse 1 m/s Melting pool fills the groove Efficient metal processing using high average power ultrafast laser John Lopez September 13th

21 Influence of scanning velocity at 10 MHz 5 m/s 4 m/s 3 m/s 2 m/s 40 µm Acceptable quality Eff m/s Burrs appear Eff m/s Melting flow appears Eff m/s Groove starts to collapse 1 m/s Thermal load rises while decreasing velocity We need higher scanning velocity & lower overlap -> Polygon scanner Overthickness 500fs 10MHz 50W 0.3J/cm² Melting pool fills the groove Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Ablation efficiency [a.u.] Ablation efficiency versus fluence Polygon 25m/s (30 passes) Depth 2.3 µm 3.6 mm 3 /min Eff. 0.07 Smooth No melting 30 pass Depth 16 µm 3.1 mm 3 /min Eff. 0.6 Smooth No melting 240 pass Fluence [J cm -2 ] 500fs 2MHz 25m/s pitch 12.

22 Ablation efficiency [a.u.] Ablation efficiency versus fluence Polygon 25m/s (30 passes) Depth 2.3 µm 3.6 mm 3 /min Eff Smooth No melting 30 pass Depth 16 µm 3.1 mm 3 /min Eff. 0.6 Smooth No melting 240 pass Fluence [J cm -2 ] 500fs 2MHz 25m/s pitch 12.5µm overlap 67.1% Ablation efficiency increases with fluence up to 3 J/cm² (70W at 2MHz) WITH a good processing quality Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Number of passes Depth 42 µm 500fs 70W 2 MHz & 25 m/s Multipass capability is proven up to 70 W with

23 Ablation efficiency [a.u.] Multipass processing capability with Polygon 10 MHz & 100 m/s 240 passes Depth 16 µm 10 MHz & 25 m/s Number of passes Depth 42 µm 500fs 70W 2 MHz & 25 m/s Multipass capability is proven up to 70 W with polygon Ablation efficiency & processing quality depend on Rep. Rate Depth 16 µm 40 µm Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Influence of repetition rate same fluence, same velocity Rep. Rate 2 MHz 4 MHz 8 MHz Average Power [W] 17.3 34.5 70 Fluence [J/cm²] 0.75 0.

08 0.08 0.08 Ablated volume [µm3/pls] 3 9 13 Ablation Efficiency [a.u.] 0.02 0.08 0.12 8 MHz 4 MHz 2 MHz Increasing rep.

24 Influence of repetition rate same fluence, same velocity Rep. Rate 2 MHz 4 MHz 8 MHz Average Power [W] Fluence [J/cm²] Velocity [m/s] Overlap [%] Nb passes Dose [J/mm] single pass Cumulative Dose [J/mm] multi pass Ablated volume [µm3/pls] Ablation Efficiency [a.u.] MHz 4 MHz 2 MHz Increasing rep. rate enhances ablation efficiency due to heat accumulation BUT leads to a bumpy surface as well 40 µm Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Influence of repetition rate same fluence, same overlap 8 MHz Rep. Rate 2 MHz 4 MHz 8 MHz Average Power [W] 17.3 34.5 70 Fluence [J/cm²] 0.75 0.

25 Influence of repetition rate same fluence, same overlap 8 MHz Rep. Rate 2 MHz 4 MHz 8 MHz Average Power [W] Fluence [J/cm²] Velocity [m/s] Overlap [%] MHz Nb passes Pulse-to-pulse delay [µs] Dose [J/mm] single pass Cumulative Dose [J/mm] multi pass Ablated volume [µm3/pls] Ablation Efficiency [a.u.] MHz Increasing the velocity and rep. rate in the same ratio enables us to maintain a good processing quality Each pulse is more efficiency due to the offset temperature Overlap seems to be the key parameter 40 µm Efficient metal processing using high average power ultrafast laser John Lopez September 13th

26 10 MHz 100m/s Overlap 74% 8 MHz 100m/s Overlap 67% Optimal overlap at 70W? Scanning Velocity (m/s) 67% 10 MHz 50m/s Overlap 87% 8 MHz 50m/s Overlap 84% no HAZ 4 MHz 50m/s Overlap 67% HAZ 10 MHz 25m/s Overlap 93% 8 MHz 25m/s Overlap 92% 4 MHz 25m/s Overlap 84% 2 MHz 25m/s Overlap 67% Pulse toultrafast pulselaser delay (µs) Efficient metal processing using high average power John Lopez September 13th

27 How long to remove 1mm 3 of Steel? Galvo scanner Polygon scanner 9 W Rep. Rate (MHz) 0.5 Galvo x 8 70 W Rep. Rate (MHz) 8 Fluence (J/cm²) 2.6 Fluence (J/cm²) 0.75 Velocity (m/s) 5 Velocity (m/s) 100 Overlap (%) 67 Polygon Overlap (%) 67 Efficiency (a.u.) 0.36 Removal Rate (mm3/min) 2.5 x2 Efficiency 0.18 Removal Rate (mm3/min) 9.6 Time (s) fs x4 Time (s) 6.3 Efficient metal processing using high average power ultrafast laser John Lopez September 13th

28 Summary Galvo scanner (up to 5 m/s) : High ablation efficiency : 0.36 at 9W (w/o burr, bumpy surface) Low ablation ablation efficiency : 0.18 at 3W (smooth, melt-free) Multipass processing : up to 9 W, drastic drop for higher P av Polygon scanner (up to 100 m/s) : Lower ablation efficiency : 0.18 at 70W (smooth, melt-free) Up-scaling has been demonstrated up to 70 W with 67% overlap Pulse to pulse delay is not a limiting factor up to 8 MHz but we expect a detrimental effect for higher rep. rate Multipass processing : up to 70W, may be more Best results 6 s/mm 3 Stainless Steel is very sensitive to parameters such as fluence, velocity, rep. rate. We expect different behavior with other metals All these results do not take into account the max. utilization rate of galvo scanner (50 to 100%) and polygon scanners (50 to 70%) that reduce the time window actually dedicated to the process Efficient metal processing using high average power ultrafast laser John Lopez September 13th

29 Acknowledgements Thank you for your kind attention Meet the author : john.lopez@u-bordeaux.fr Efficient metal processing using high average power ultrafast laser John Lopez September 13th

Efficient Metal Processing Using High Average Power Ultrafast Laser

Efficient Metal Processing Using High Average Power Ultrafast Laser John Lopez *1, Konstantin Mishchik *2, Girolamo Mincuzzi *3, Eric Audouard *2, Eric Mottay *2 and Rainer Kling *3 *1 UNIV BORDEAUX, CELIA